A preliminary study on computerized lesion localization in MR mammography using 3D nMITR maps, multilayer cellular neural networks, and fuzzy c-partitioning
A preliminary study on computerized lesion localization in MR mammography using 3D nMITR maps, multilayer cellular neural networks, and fuzzy c-partitioning
复制标题
DOI：
10.1118/1.2805477
复制
发表时间：
2008-01-01
期刊：
影响因子：
3.8
通讯作者：
Ucan, Osman Nuri
中科院分区：
文献类型：
--
作者：
Ertas, Gokhan;Gulcur, H. Ozcan;Ucan, Osman Nuri
细胞神经网络（CNN）是具有学习能力的大规模并行细胞结构。它们可用于有效地且几乎真实的实现复杂的图像处理应用。在这项初步研究中，我们提出了一种基于CNN的新型，强大和全自动化系统，以促进对比增强MR乳腺X射线摄影中的病变定位，这是一项艰巨的任务，需要处理大量图像，并注意微小的细节。该数据集由1170个切片组成，包含39名患者的1个对比前和5个对比后双侧轴向MR乳腺X线片，其中37个恶性和39个良性肿块病变使用1.5 T MR扫描仪采集，参数如下：3D FLASH序列，TR/TE 9.80/4.76 ms，翻转角25 °，层厚2.5 mm，平面分辨率0.625 x 0.625 mm（2）。该集合的600个切片（21个良性病变和25个恶性病变）用于训练CNN;其余数据用于测试目的。首先使用级联连接的四个2D CNN从造影前图像中分割感兴趣的乳房区域，专门设计用于最大限度地减少由于肌肉，心脏，肺和胸腔引起的错误检测。为了识别欺骗性增强区域，计算分割的乳房的3D nMITR图并将其转换为二进制形式。在此过程中，具有低程度增强的组织被丢弃。为了增强病变，该二进制图像由3D CNN处理，该CNN具有由三层11 x 11细胞组成的控制模板和模糊c分区输出函数。基于体积和3D偏心率特征从训练数据集中凭经验提取的一组决策规则用于做出最终决策并定位病变。分割算法表现良好，具有高平均精度、高真阳性体积分数和低假阳性体积分数，总体性能分别为0.93 +/- 0.05、0.96 +/- 0.04和0.03 +/- 0.05（训练：0.93 +/- 0.04、0.94 +/- 0.04和0.02 +/- 0.03;测试：0.93 +/- 0.05、0.97 +/- 0.03和0.05 +/- 0.06）。系统的病变检测性能相当令人满意;对于训练数据集，最大检测灵敏度为100%，假阳性检测为0.28/病变，0.09/切片，0.65/例;对于测试数据集，最大检测灵敏度为97%，假阳性检测为0.43/病变，0.11/切片，0.68/例。平均而言，对于99%的检测灵敏度，系统的总体性能为0.34/病变、0.10/切片和0.67/病例。介绍的系统不需要有关乳房解剖结构的先验信息，它是强大的，非常有效的检测乳腺病变。CNN、模糊c分区、体积和3D偏心率标准的使用减少了由于高度增强的血管、乳头和正常实质引起的伪影以及由于过度分割引起的胸壁血管化组织的伪影而导致的假阳性检测。我们希望该系统将有助于乳腺检查，改善病变的定位，并减少不必要的乳房切除术，特别是由于错过多中心病变和几乎实时的处理速度可实现的直接硬件实现将开辟新的临床应用，如可行的准自动MR引导活检和收购额外的对比后病变图像，以改善形态特征。(c)2008年美国医学物理学家协会。
Cellular neural networks (CNNs) are massively parallel cellular structures with learning abilities. They can be used to realize complex image processing applications efficiently and in almost real time. In this preliminary study, we propose a novel, robust, and fully automated system based on CNNs to facilitate lesion localization in contrast-enhanced MR mammography, a difficult task requiring the processing of a large number of images with attention paid to minute details. The data set consists of 1170 slices containing one precontrast and five postcontrast bilateral axial MR mammograms from 39 patients with 37 malignant and 39 benign mass lesions acquired using a 1.5 Tesla MR scanner with the following parameters: 3D FLASH sequence, TR/TE 9.80/4.76 ms, flip angle 25 degrees, slice thickness 2.5 mm, and 0.625 x 0.625 mm(2) in-plane resolution. Six hundred slices with 21 benign and 25 malignant lesions of this set are used for training the CNNs; the remaining data are used for test purposes. The breast region of interest is first segmented from precontrast images using four 2D CNNs connected in cascade, specially designed to minimize false detections due to muscles, heart, lungs, and thoracic cavity. To identify deceptively enhancing regions, a 3D nMITR map of the segmented breast is computed and converted into binary form. During this process tissues that have low degrees of enhancements are discarded. To boost lesions, this binary image is processed by a 3D CNN with a control template consisting of three layers of 11 x 11 cells and a fuzzy c-partitioning output function. A set of decision rules extracted empirically from the training data set based on volume and 3D eccentricity features is used to make final decisions and localize lesions. The segmentation algorithm performs well with high average precision, high true positive volume fraction, and low false positive volume fraction with an overall performance of 0.93 +/- 0.05, 0.96 +/- 0.04, and 0.03 +/- 0.05, respectively (training: 0.93 +/- 0.04, 0.94 +/- 0.04, and 0.02 +/- 0.03; test: 0.93 +/- 0.05, 0.97 +/- 0.03, and 0.05 +/- 0.06). The lesion detection performance of the system is quite satisfactory; for the training data set the maximum detection sensitivity is 100% with false-positive detections of 0.28/lesion, 0.09/slice, and 0.65/case; for the test data set the maximum detection sensitivity is 97% with false-positive detections of 0.43/lesion, 0.11/slice, and 0.68/case. On the average, for a detection sensitivity of 99%, the overall performance of the system is 0.34/lesion, 0.10/slice, and 0.67/case. The system introduced does not require prior information concerning breast anatomy; it is robust and exceptionally effective for detecting breast lesions. The use of CNNs, fuzzy c-partitioning, volume, and 3D eccentricity criteria reduces false-positive detections due to artifacts caused by highly enhanced blood vessels, nipples, and normal parenchyma and artifacts from vascularized tissues in the chest wall due to oversegmentation. We hope that this system will facilitate breast examinations, improve the localization of lesions, and reduce unnecessary mastectomies, especially due to missed multicentric lesions and that almost real-time processing speeds achievable by direct hardware implementations will open up new clinical applications, such as making feasible quasi-automated MR-guided biopsies and acquisition of additional postcontrast lesion images to improve morphological characterizations.(c) 2008 American Associationof Physicists in Medicine.